Intelligent all-terrain embankment hidden danger rapid diagnosis crawler and operation method

Abstract

The invention discloses an intelligent all-terrain embankment hidden danger rapid diagnosis tracked vehicle and an operation method. The tracked vehicle comprises a vehicle body, a four-wheel independent suspension track structure, a geological radar detection module, a transient electromagnetic detection module, a multi-mode navigation system, an intelligent control module and a remote control platform. All-terrain path planning, autonomous obstacle avoidance and cooperative detection of the intelligent all-terrain embankment hidden danger rapid diagnosis crawler are realized based on a multi-mode navigation system, an intelligent control module and a remote control platform; through the technologies of a four-wheel independent suspension track structure, dual-mode cooperative detection, real-time data transmission and inversion and the like, the detection range is remarkably expanded, the hidden danger recognition precision and reliability are improved, the time interval between field diagnosis and emergency response is shortened, the bottleneck problems in the aspects of terrain adaptability, detection precision, timeliness and the like in the prior art are solved, and the potential safety hazard detection method is suitable for large-scale popularization and application. And the comprehensiveness, the accuracy and the emergency response speed of dike hidden danger diagnosis are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of levee engineering detection, and specifically to an intelligent all-terrain levee hidden danger rapid diagnosis tracked vehicle and an operation method thereof. Background Art

[0002] As the core barrier for flood control safety, the accurate detection and rapid diagnosis of hidden dangers inside levees are directly related to major livelihood safety. Although current geophysical exploration technologies have been applied in levee detection, there are still significant technical bottlenecks, resulting in low hidden danger identification efficiency and insufficient result reliability. The specific manifestations are as follows key problems: First, the automation level of detection equipment is low and the terrain adaptability is poor: Existing detection equipment mostly relies on manual operation, with low efficiency and discontinuous data collection, and it is difficult to meet the high-intensity inspection requirements during the flood season. Conventional detection vehicles are limited by the wheeled chassis structure and power performance, and it is difficult to adapt to complex terrains such as steep and muddy levee slopes, resulting in insufficient detection coverage and a significant risk of missed detections. Second, the limitations of traditional detection methods are prominent and the detection accuracy is low: The physical properties of levee media change dynamically, and the physical property parameters interfere with each other. Single geophysical methods are easily interfered by geological conditions, and the problem of multiple solutions is serious; Existing inversion methods lack a multi-source data fusion mechanism, resulting in too large an error in the determination of cavity hidden dangers. Third, the on-site judgment ability is lacking and the emergency response is lagging: The detection data of current geophysical exploration equipment needs to be processed offline later, and it takes too long from collection to the generation of the result report, and it cannot support real-time decision-making during the flood season.

[0003] Existing technologies have proposed multi-modal detection schemes, but their vehicle-mounted designs are limited to flat terrains and do not solve the data fusion problem. There is an urgent need to achieve technological breakthroughs in the following directions: ① Develop an intelligent all-terrain detection platform to break through the limitations of equipment on complex terrains; ② Construct a multi-physical field joint inversion model to improve the accuracy of hidden danger identification; ③ Establish a remote transmission architecture to achieve on-site rapid diagnosis and emergency response. Summary of the Invention

[0004] Aiming at the deficiencies of the existing technology, the present invention provides an intelligent all-terrain levee hidden danger rapid diagnosis tracked vehicle and an operation method thereof, and its purpose is to improve the hidden danger detection operation ability in complex environments through technologies such as a four-wheel independent suspension tracked structure, dual-modal collaborative detection, real-time data transmission and inversion, and achieve rapid diagnosis and emergency response of hidden dangers.

[0005] To achieve the above purpose, the present invention provides the following technical solutions: An intelligent all-terrain levee hidden danger rapid diagnosis tracked vehicle, comprising: A vehicle body; A four-wheel independent suspension tracked structure, which is arranged at the bottom of the vehicle body and is used to drive the vehicle body to move; A power system, which is arranged inside the vehicle body and is used to transmit power to the four-wheel independent suspension tracked structure; The ground penetrating radar detection module is installed on the bottom plate of the vehicle body abdomen and is used to collect ground penetrating radar data for geological structure detection; The transient electromagnetic detection module is installed on the vehicle body and is used to collect transient electromagnetic data for geological structure detection; The multi-modal navigation system is set on the vehicle body and is used to generate an environmental map in real time, take real-time photos, and detect obstacles; The intelligent control module is set inside the vehicle body and is used to receive signals and respond to the signals; Responding to the signals includes vehicle motion control based on the multi-modal navigation system and transmission of the collected data; The collected data includes ground penetrating radar data for geological structure detection and transient electromagnetic data for geological structure detection; The remote control platform is used to control the intelligent all-terrain levee hidden danger rapid diagnosis tracked vehicle, receive the collected data, preprocess the collected data, and then perform collaborative inversion calculation to output the hidden danger area; Based on the multi-modal navigation system, intelligent control module and remote control platform, the all-terrain path planning-autonomous obstacle avoidance-collaborative detection of the intelligent all-terrain levee hidden danger rapid diagnosis tracked vehicle is realized.

[0006] Furthermore, the multi-modal navigation system includes: The integrated RTK-GNSS differential positioning module is installed inside the vehicle body and is used to provide global coordinates; The lidar is installed inside the vehicle body and is used to scan the environment in real time to generate three-dimensional point cloud data; The binocular vision camera is installed at the front end of the vehicle body and is used to take real-time pictures; The IMU inertial navigation unit is installed inside the vehicle body and is used to realize vehicle body attitude measurement, navigation positioning and motion monitoring; The wheel speed encoder is installed on the wheel set of the four-wheel independent suspension tracked structure and is used to measure the rotational speed of the wheel set.

[0007] Furthermore, the intelligent control module includes: The wireless communication unit is set inside the vehicle body and is used to receive or transmit signals; The central control unit is set inside the vehicle body. The central control unit is communicatively connected to the wireless communication unit and is used to receive the signals transmitted by the wireless communication unit and make responses; The wireless communication unit includes a multi-band router and a 5G communication module set on the top of the vehicle body, and the signals are received and transmitted through the multi-band router and the 5G communication module.

[0008] Further, the remote control platform integrates an intelligent vehicle remote control system, a data fusion processing system, and a display device; the intelligent vehicle remote control system is used to control the intelligent all-terrain levee hidden danger rapid diagnosis tracked vehicle; the data fusion processing system is used to preprocess the collected data, and then perform collaborative inversion calculations to output the hidden danger area.

[0009] Further, the central control unit is electrically connected to the geological radar detection module, the transient electromagnetic detection module, the integrated RTK-GNSS differential positioning module, the lidar, the binocular vision camera, the IMU inertial navigation unit, and the wheel speed encoder.

[0010] An operation method of an intelligent all-terrain levee hidden danger rapid diagnosis tracked vehicle includes the following steps: Step S1: The intelligent all-terrain levee hidden danger rapid diagnosis tracked vehicle arrives at the operation area of the target levee section; Step S2: The intelligent all-terrain levee hidden danger rapid diagnosis tracked vehicle performs all-terrain path planning - autonomous obstacle avoidance - collaborative detection on the operation area of the target levee section; Step S3: Transmit the geological structure detection radar data and the geological structure detection transient electromagnetic data collected by detection to the data fusion processing system for processing. When it is detected that there are hidden dangers in the operation area of the target levee section, output the hidden danger position coordinates through the multimodal navigation system, and generate a hidden danger distribution map; Step S4: Push the hidden danger position coordinates and the hidden danger distribution map to the command center. The command center determines the hidden danger occurrence area based on the pushed hidden danger position coordinates and the hidden danger distribution map, and timely conducts emergency treatment on the levee sections with hidden dangers.

[0011] Further, the specific process of the intelligent all-terrain levee hidden danger rapid diagnosis tracked vehicle performing all-terrain path planning - autonomous obstacle avoidance - collaborative detection on the operation area of the target levee section is as follows: The intelligent all-terrain levee hidden danger rapid diagnosis tracked vehicle scans the environment in real time through the lidar to generate three-dimensional point cloud data, combines the global coordinates provided by the integrated RTK-GNSS differential positioning module to generate an electronic map, and transmits the electronic map to the remote control platform through the intelligent control module for path planning; The intelligent all-terrain levee hidden danger rapid diagnosis tracked vehicle travels according to the planned path, generates the pose of the intelligent all-terrain levee hidden danger rapid diagnosis tracked vehicle during the driving process through the integrated RTK-GNSS differential positioning module, the IMU inertial navigation unit, and the wheel speed encoder, and compensates for the positioning drift caused by the bump of the intelligent all-terrain levee hidden danger rapid diagnosis tracked vehicle; detects obstacles and slope angles in the driving forward direction through the lidar and the binocular vision camera; Based on the dynamic window method, the feasible speed space of the intelligent all-terrain crawler vehicle for rapid diagnosis of dike hidden dangers is evaluated in real time. Based on the feasible speed space, obstacles, and slope inclination, the driving path of the intelligent all-terrain crawler vehicle for rapid diagnosis of dike hidden dangers is dynamically adjusted, and the optimal path is selected; The movement / stop signal of the intelligent all-terrain crawler vehicle for rapid diagnosis of dike hidden dangers is collected through a wheel speed encoder. Based on the movement / stop signal, the geological radar detection module and the transient electromagnetic detection module are controlled to synchronously collect / stop data acquisition work according to the optimal path, and the collected data is transmitted to the remote control platform through the wireless communication unit.

[0012] Furthermore, the dynamic window method selects the optimal speed pair of the intelligent all-terrain crawler vehicle for rapid diagnosis of dike hidden dangers through a multi-objective weighted evaluation function, including the linear velocity and the angular velocity , expressed as: ; In the formula, represents the cost function, which is used to evaluate the comprehensive performance of the vehicle when driving at the linear velocity and the angular velocity ; represents the heading angle term; represents the obstacle distance term; represents the speed term; represents the weight coefficient of the heading angle term, which is used to adjust the proportion of the heading angle term in the cost function; represents the weight coefficient of the obstacle distance term, which is used to adjust the proportion of the obstacle distance term in the cost function; represents the weight coefficient of the speed term, which is used to adjust the proportion of the speed term in the cost function.

[0013] Furthermore, the detected and collected data is transmitted to the data fusion processing system for processing. When hidden dangers are detected in the operation area of the target dike section, the specific process of outputting the hidden danger location coordinates through the multi-modal navigation system and generating a hidden danger distribution map is as follows: Collect the geological structure detection radar data and the geological structure detection transient electromagnetic data, and preprocess the collected geological structure detection radar data and the geological structure detection transient electromagnetic data; According to prior information or experience, set the initial resistivity and the dielectric constant distribution. Based on the initial resistivity and the dielectric constant distribution, construct a forward model; According to the resistivity distribution in the forward model, generate forward transient electromagnetic response data; according to the dielectric constant Distribution to generate forward ground penetrating radar response data; Compare the forward transient electromagnetic response data with the collected transient electromagnetic data for geological structure detection, and calculate the electromagnetic data fitting error ; Compare the forward ground penetrating radar response data with the collected ground penetrating radar data for geological structure detection, and calculate the radar data fitting error ; According to the current resistivity and dielectric constant distribution, calculate the cross-gradient constraint term : ; In the formula, is the gradient operator; Combine , and to construct the objective function of the joint inversion model based on the Bayesian framework : ; In the formula, represents the model parameters, including resistivity and dielectric constant ; α, β, λ are the weight coefficients of each item; Adopt the gradient descent method to optimize the inversion model parameters by minimizing the objective function : ; In the formula, represents the model parameter vector after the (k + 1)-th iteration; represents the learning rate; After each iteration, check whether the change rate of the objective function is less than the preset change rate or whether the number of iterations reaches the preset maximum value. If so, stop the iteration; otherwise, continue to iterate and update the model parameters; Combine the optimized model parameters with the digital elevation model to generate a hidden danger distribution map and perform three-dimensional visualization annotation on the hidden danger area.

[0014] Compared with the existing technologies, the present invention has the following beneficial effects:

[0015] (1)By integrating technologies such as four-wheel independent suspension track structure, dual-modal collaborative detection, real-time data transmission and inversion onto the vehicle body, the present invention significantly expands the detection range of the tracked vehicle, improves the accuracy and reliability of hidden danger identification, shortens the time interval between on-site diagnosis and emergency response, systematically solves the bottleneck problems of the prior art in terms of terrain adaptability, detection accuracy, timeliness, etc., greatly enhances the comprehensiveness, accuracy of dike hidden danger diagnosis and the emergency response speed, and has remarkable technological advancement and engineering practical value.

[0016] (2)Through the multi-modal navigation system, intelligent control module and remote control platform set on the tracked vehicle, the present invention can control the tracked vehicle for path planning - autonomous obstacle avoidance - collaborative detection, realizing intelligent and automated detection operation of dike hidden dangers in all terrains under complex environments, reducing the consumption of manpower and material resources; by collecting the complementary advantages of multi-physical field data, the efficiency of hidden danger detection operation and the accuracy of results are significantly improved.

[0017] (3)The present invention can push the hidden danger location coordinates and hidden danger distribution maps to the rear command center for experts and emergency rescue teams to master the dike conditions, quickly determine the hidden danger occurrence areas, and promptly conduct emergency treatment on the dike sections with hidden dangers, realizing the rapid diagnosis and emergency response of dike hidden dangers. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is the external view of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle of the present invention.

[0019] Figure 2 It is the internal structure diagram of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle of the present invention.

[0020] Figure 3 It is the internal end structure diagram of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle of the present invention.

[0021] Figure 4 It is the method flow chart of the present invention.

[0022] Figure 5 It is the on-site real-time operation schematic diagram of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle of the present invention.

[0023] In the figure, 110, four-wheel independent suspension track structure; 120, lithium battery pack; 210, radar antenna; 310, electromagnetic coil; 330, flip-up support arm; 410, integrated RTK-GNSS differential positioning module; 420, lidar; 430, binocular vision camera; 440, IMU inertial navigation unit; 450, wheel speed encoder; 511, multi-band router; 512, 5G communication module; 520, central control unit. DETAILED DESCRIPTION OF THE INVENTION

[0024] Such asFigures 1-3 As shown in the figure, the present invention provides a technical solution: an intelligent all-terrain levee hidden danger rapid diagnosis crawler vehicle, including:

[0025] A vehicle body.

[0026] A four-wheel independent suspension crawler structure 110, which is arranged at the bottom of the vehicle body and is used to drive the vehicle body to move; the four-wheel independent suspension crawler structure 110 can adapt to complex terrains with a slope ≤ 30°.

[0027] A power system, which is arranged inside the vehicle body and is used to transmit power to the four-wheel independent suspension crawler structure.

[0028] A geological radar detection module, which is installed on the bottom plate of the vehicle body abdomen and is installed close to the ground, and is used to collect geological structure detection radar data.

[0029] A transient electromagnetic detection module, which is installed on the vehicle body and is used to collect geological structure detection transient electromagnetic data.

[0030] A multimodal navigation system, which is arranged on the vehicle body and is used to generate a high-precision environmental map in real time and plan a driving path, photograph the surrounding environment, detect obstacles and avoid them autonomously during the operation, so as to realize autonomous operation in a complex environment.

[0031] An intelligent control module, which is arranged inside the vehicle body and is used to receive signals and respond to the signals (control the vehicle movement based on the multimodal navigation system and transmit the collected data); the collected data includes geological structure detection radar data and geological structure detection transient electromagnetic data.

[0032] Among them, the radar antenna 210 of the geological radar detection module is installed on the top of the vehicle body. The radar antenna 210 can select the GX160 high-dynamic shielding antenna of MALA Company in Sweden, and the transmission cable of the geological radar detection module can select the RG-214 / U double-shielded cable.

[0033] Among them, the electromagnetic coil 310 of the transient electromagnetic detection module is installed at the front end of the vehicle body through a flip support arm 330; the electromagnetic coil 310 can adopt a large-current multi-turn loop coil with a diameter of 60 cm; the communication cable of the transient electromagnetic detection module can adopt a high-protection type cable of the LMR-400-DB model; the flip support arm 330 can adopt a carbon fiber material.

[0034] Among them, the multimodal navigation system includes:

[0035] Integrated RTK-GNSS differential positioning module 410, which is installed at the rear end inside the vehicle body and is used to provide global coordinates; the integrated RTK-GNSS differential positioning module 410 can be selected as the UM960 model of U-blox.

[0036] LiDAR 420, which is installed inside the vehicle body and is used to scan the environment in real time to generate three-dimensional point cloud data; the LiDAR 420 can be selected as the Fashi LiDAR S30.

[0037] Binocular vision camera 430, which is installed at the front end of the vehicle body and is used to capture real-time images; the binocular vision camera 430 can be selected as the Stereolabs ZED 2 binocular vision stereo camera.

[0038] IMU inertial navigation unit 440, which is installed inside the vehicle body and is used to achieve vehicle body attitude measurement, navigation positioning and motion monitoring; the IMU inertial navigation unit 440 can be selected as the VectorNav VN-100 inertial navigation system.

[0039] Wheel speed encoder 450, which is installed on the wheel set of the four-wheel independent suspension track structure 110 and is used to measure the rotational speed of the wheel set; the wheel speed encoder 450 can be selected as the Omron E6B2-CWZ3E incremental rotary encoder.

[0040] Among them, the intelligent control module includes:

[0041] Wireless communication unit, which is set inside the vehicle body and is used to receive or transmit signals; the wireless communication unit can be selected as the SIM800L module.

[0042] Central control unit 520, which is set inside the vehicle body. The central control unit 520 is communicatively connected to the wireless communication unit and is used to receive the signals transmitted by the wireless communication unit and make responses; the central control unit 520 can be selected as the Siemens S7-1200PLC.

[0043] The wireless communication unit includes a multi-band router 511 and a 5G communication module 512 arranged on the top of the vehicle body. Wireless connection is carried out through the multi-band router 511 and the 5G communication module 512 to achieve signal reception and transmission; the multi-band router 511 can be selected as the TP-Link Archer AX6000 router; the 5G communication module 512 can adopt the Qualcomm Snapdragon X55 chip.

[0044] Among them, it also includes a lithium battery pack 120 for powering the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle, with a battery life of ≥ 8 hours, an operating speed of 5.4 km / h, and an operating efficiency of 5000 m 2 / h, meeting the detection efficiency requirements.

[0045] Among them, it also includes a remote control platform for controlling the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle.

[0046] The remote control platform integrates an intelligent vehicle remote control system, a data fusion processing system, and a display device; the intelligent vehicle remote control system is used to control the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle; the data fusion processing system is used to preprocess the collected data, and then perform collaborative inversion calculations to output the hidden danger area; the display device is used to display the status information of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle, an electronic map marking the position of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle, real-time collected data, and hidden danger detection results, etc.; the display device also supports interactive operations; the intelligent vehicle remote control system can select a Raspberry Pi 4 computer; the data fusion processing system can select an NVIDIA Jetson embedded system.

[0047] The central control unit 520 is electrically connected to the geological radar detection module, the transient electromagnetic detection module, the integrated RTK-GNSS differential positioning module 410, the lidar 420, the binocular vision camera 430, the IMU inertial navigation unit 440, and the wheel speed encoder 450.

[0048] The vehicle body is made of carbon steel, and the four-wheel independent suspension track structure 110 is equipped with anti-skid rubber tracks and a hydraulic leveling module for leveling the four-wheel independent suspension track structure 110.

[0049] Among them, the all-terrain path planning - autonomous obstacle avoidance - collaborative detection of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle is realized based on the multi-modal navigation system, the intelligent control module, and the remote control platform.

[0050] As Figure 4 shown, an operation method of an intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle includes the following steps:

[0051] Step S1: The intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle arrives at the operation area of the target dike section, as Figure 5 shown.

[0052] Step S2: The intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle performs all-terrain path planning - autonomous obstacle avoidance - collaborative detection on the operation area of the target dike section.

[0053] Step S3: Transmit the geological structure detection radar data and the geological structure detection transient electromagnetic data detected and collected to the data fusion processing system for processing. When it is detected that there are hidden dangers in the operation area of the target dike section, output the coordinates of the hidden danger location through the multi-modal navigation system, and generate a hidden danger distribution map.

[0054] Step S4: Push the coordinates of the hidden danger location and the hidden danger distribution map to the command center. The command center determines the hidden danger occurrence area based on the pushed coordinates of the hidden danger location and the hidden danger distribution map, and promptly conducts emergency treatment on the dike section with hidden dangers.

[0055] Among them, the specific process of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle for all-terrain path planning - autonomous obstacle avoidance - collaborative detection in the operation area of the target dike section is as follows:

[0056] 1. The intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle scans the environment in real time through the lidar 420 to generate three-dimensional point cloud data. Combining the global coordinates provided by the integrated RTK-GNSS differential positioning module 410, a high-precision electronic map is generated. The electronic map is transmitted to the remote control platform through the intelligent control module, and path planning (setting parameters such as detection routes, detection intervals, and speeds) is carried out.

[0057] 2. The intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle travels according to the planned path. The pose of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle during driving is generated through the integrated RTK-GNSS differential positioning module 410, the IMU inertial navigation unit 440, and the wheel speed encoder 450, and the positioning drift caused by the bump of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle is compensated; obstacles and slope angles in the driving forward direction are detected through the lidar 420 and the binocular vision camera 430; among them, the lidar 420 detects obstacles at higher positions (height > 20 cm), and the binocular vision camera 430 detects low obstacles (height < 20 cm) and slope angles.

[0058] 3. Based on the dynamic window approach (DWA), the feasible speed space of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle is evaluated in real time. Based on the feasible speed space, obstacles, and slope angles, the driving path of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle is dynamically adjusted, and the optimal path is selected to ensure that the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle travels along the planned path.

[0059] Among them, the dynamic window approach (DWA) selects the optimal speed pair (linear speed and angular speed ) of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle through a multi-objective weighted evaluation function, which is expressed as: ; In the formula, represents the cost function, which is used to evaluate the comprehensive performance of the vehicle running at different speed combinations (linear speed and angular speed ). The smaller the value, the better the speed combination; represents the heading angle term; represents the obstacle distance term; represents the speed term; represents the weight coefficient of the heading angle term, which is used to adjust the proportion of the heading angle term in the cost function; represents the weight coefficient of the obstacle distance term, which is used to adjust the proportion of the obstacle distance term in the cost function; represents the weight coefficient of the speed term, which is used to adjust the proportion of the speed term in the cost function; in this embodiment, , , .

[0060] 4. The movement / stop signal of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle is collected by the wheel speed encoder 450. Based on the movement / stop signal, the geological radar detection module and the transient electromagnetic detection module are controlled to synchronously perform / stop data collection work according to the optimal path (instructions can also be issued manually through the intelligent vehicle remote control system). The collected data is transmitted to the remote control platform through the wireless communication unit; the operation time sequences of the geological radar detection module and the transient electromagnetic detection module are aligned by GPS timing (time deviation ≤ 1 ms); the data timestamp is bound to the positioning coordinates of the integrated RTK-GNSS differential positioning module 410 (the installation spacing between the geological radar detection module and the transient electromagnetic detection module ≥ 2 m) to ensure spatial consistency.

[0061] Among them, the detected and collected data (geological structure detection radar data and geological structure detection transient electromagnetic data) is transmitted to the data fusion processing system for processing. When it is detected that there are hidden dangers in the operation area of the target dike section, the specific process of outputting the hidden danger position coordinates through the multi-modal navigation system and generating the hidden danger distribution map is as follows:

[0062] 1. Data collection and preprocessing.

[0063] Collect the geological structure detection radar data and the geological structure detection transient electromagnetic data, and preprocess the collected geological structure detection radar data and the geological structure detection transient electromagnetic data, including operations such as time synchronization, spatial registration, and noise suppression.

[0064] 2. Initial model establishment.

[0065] According to prior information or experience, set the initial resistivity and relative permittivity distribution. Based on the initial resistivity and dielectric constant Construct a forward model for the distribution to generate forward response data.

[0066] 3. Forward calculation.

[0067] According to the resistivity in the forward model distribution, generate forward transient electromagnetic response data; according to the dielectric constant in the forward model distribution, generate forward ground penetrating radar response data.

[0068] 4. Calculate the fitting error.

[0069] Compare the forward transient electromagnetic response data with the collected transient electromagnetic data for geological structure detection, and calculate the fitting error of the electromagnetic data ; compare the forward ground penetrating radar response data with the collected ground penetrating radar data for geological structure detection, and calculate the fitting error of the radar data .

[0070] 5. Cross-gradient constraint calculation.

[0071] According to the current resistivity and dielectric constant distribution, calculate the cross-gradient constraint term : ; In the formula, is the gradient operator.

[0072] 6. Construct the objective function.

[0073] Combine , and to construct the objective function of the joint inversion model based on the Bayesian framework : ; In the formula, represents the model parameters, including resistivity and dielectric constant ; α = 0.6, β = 0.3, λ = 0.1 are the weight coefficients of each item, which are determined by cross-validation of the model data.

[0074] 7. Optimize the model parameters.

[0075] Adopt the gradient descent method to optimize the inversion model parameters by minimizing the objective function , and update the inversion model parameters iteratively according to the following formula: ; In the formula, denotes the model parameter vector after the (k + 1)-th iteration; denotes the learning rate.

[0076] After each iteration, check whether the change rate of the objective function is less than 1% or the number of iterations has reached the maximum value (e.g., 50 times). If the termination condition is met, stop the iteration; otherwise, continue to iterate and update the model parameters.

[0077] 8. Output the inversion result.

[0078] Combine the optimized model parameters (resistivity and dielectric constant ) with the digital elevation model (DEM) to generate a hidden danger distribution map and perform three-dimensional visualization annotation on the hidden danger areas.

[0079] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent all-terrain crawler vehicle for rapid diagnosis of hidden dangers in dikes, characterized in that, Including: Vehicle body; Four-wheel independent suspension track structure, which is arranged at the bottom of the vehicle body and used to drive the vehicle body to move; Power system, which is arranged inside the vehicle body and used to transmit power to the four-wheel independent suspension track structure; Ground penetrating radar detection module, which is installed on the bottom plate of the vehicle body abdomen and used to collect ground penetrating radar data for geological structure detection; Transient electromagnetic detection module, which is installed on the vehicle body and used to collect transient electromagnetic data for geological structure detection; Multi-modal navigation system, which is arranged on the vehicle body and used to generate an environmental map in real time, take real-time photos, and detect obstacles; Intelligent control module, which is arranged inside the vehicle body and used to receive signals and respond to the signals; Responding to the signals includes controlling the vehicle movement based on the multi-modal navigation system and transmitting the collected data; The collected data includes ground penetrating radar data for geological structure detection and transient electromagnetic data for geological structure detection; Remote control platform, which is used to control the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle, receive the collected data, preprocess the collected data, and then perform collaborative inversion calculation to output the hidden danger area; Based on the multi-modal navigation system, intelligent control module and remote control platform, realize the all-terrain path planning - autonomous obstacle avoidance - collaborative detection of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle.

2. The intelligent all-terrain crawler vehicle for rapid diagnosis of dike hidden dangers according to claim 1, characterized in that: The multi-modal navigation system includes: Integrated RTK-GNSS differential positioning module, which is installed inside the vehicle body and used to provide global coordinates; LiDAR, which is installed inside the vehicle body and used to scan the environment in real time and generate three-dimensional point cloud data; Binocular vision camera, which is installed at the front end of the vehicle body and used to take real-time pictures; IMU inertial navigation unit, which is installed inside the vehicle body and used to realize vehicle body attitude measurement, navigation positioning and motion monitoring; Wheel speed encoder, which is installed on the wheel set of the four-wheel independent suspension track structure and used to measure the rotational speed of the wheel set.

3. An intelligent all-terrain crawler vehicle for rapid diagnosis of dike hidden dangers according to claim 2, characterized in that: The intelligent control module includes: Wireless communication unit, which is arranged inside the vehicle body and used to receive or transmit signals; Central control unit, which is arranged inside the vehicle body. The central control unit is communicatively connected to the wireless communication unit and used to receive the signals transmitted by the wireless communication unit and make responses; The wireless communication unit includes a multi-band router and a 5G communication module arranged on the top of the vehicle body, and realizes receiving and transmitting signals through the multi-band router and the 5G communication module.

4. An intelligent all-terrain crawler vehicle for rapid diagnosis of dike hidden dangers according to claim 3, characterized in that: The remote control platform integrates an intelligent vehicle remote control system, a data fusion processing system and a display device; The intelligent vehicle remote control system is used to control the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle; The data fusion processing system is used to preprocess the collected data, then perform collaborative inversion calculation, and output the hidden danger area.

5. An intelligent all-terrain track vehicle for rapid diagnosis of hidden dangers of dikes according to claim 4, characterized in that: The central control unit is electrically connected to the ground penetrating radar detection module, the transient electromagnetic detection module, the integrated RTK-GNSS differential positioning module, the lidar, the binocular vision camera, the IMU inertial navigation unit, and the wheel speed encoder.

6. An operation method of the intelligent all-terrain levee hidden danger rapid diagnosis crawler vehicle according to any one of claims 1-5, characterized in that, It includes the following steps: Step S1: The intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle arrives at the operation area of the target dike section; Step S2: The intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle conducts all-terrain path planning - autonomous obstacle avoidance - collaborative detection on the operation area of the target dike section; Step S3: Transmit the ground penetrating radar data and transient electromagnetic data of the geological structure detection collected by detection to the data fusion processing system for processing. When it is detected that there are hidden dangers in the operation area of the target dike section, output the coordinate of the hidden danger position through the multi-modal navigation system, and generate a hidden danger distribution map; Step S4: Push the coordinate of the hidden danger position and the hidden danger distribution map to the command center. The command center determines the hidden danger occurrence area based on the pushed coordinate of the hidden danger position and the hidden danger distribution map, and promptly conducts emergency treatment on the dike section with hidden dangers.

7. The operation method of an intelligent all-terrain track vehicle for rapid diagnosis of dike hidden dangers according to claim 6, characterized in that: The specific process of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle conducting all-terrain path planning - autonomous obstacle avoidance - collaborative detection on the operation area of the target dike section is as follows: The intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle scans the environment in real time through the lidar to generate three-dimensional point cloud data. Combining with the global coordinates provided by the integrated RTK-GNSS differential positioning module, generate an electronic map, and transmit the electronic map to the remote control platform through the intelligent control module, and conduct path planning; The intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle travels according to the planned path, generates the pose during the travel of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle through the integrated RTK-GNSS differential positioning module, the IMU inertial navigation unit, and the wheel speed encoder, and compensates for the positioning drift caused by the bump of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle; Detect obstacles and slope angles in the driving forward direction through the lidar and the binocular vision camera; Based on the dynamic window method, real-time evaluate the feasible speed space of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle. Based on the feasible speed space, obstacles, and slope angles, dynamically adjust the driving path of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle, and select the optimal path; Collect the moving / stopping signal of the intelligent all-terrain dike hidden danger rapid diagnosis tracked vehicle through the wheel speed encoder. Based on the moving / stopping signal, control the ground penetrating radar detection module and the transient electromagnetic detection module to synchronously conduct / stop data collection work according to the optimal path, and the collected data is transmitted to the remote control platform through the wireless communication unit.

8. The operation method of an intelligent all-terrain levee hidden danger rapid diagnosis crawler vehicle according to claim 7, characterized in that: The dynamic window method selects the optimal speed pair of the intelligent all-terrain crawler vehicle for rapid diagnosis of dike hidden dangers through a multi-objective weighted evaluation function, including the linear velocity and the angular velocity , which is expressed as: ； In the formula, represents the cost function, which is used to evaluate the comprehensive performance of the vehicle running at the linear velocity and the angular velocity ; represents the heading angle term; represents the obstacle distance term; represents the velocity term; represents the weight coefficient of the heading angle term, which is used to adjust the proportion of the heading angle term in the cost function; represents the weight coefficient of the obstacle distance term, which is used to adjust the proportion of the obstacle distance term in the cost function; represents the weight coefficient of the velocity term, which is used to adjust the proportion of the velocity term in the cost function.

9. The operation method of an intelligent all-terrain levee hidden danger rapid diagnosis crawler vehicle according to claim 8, characterized in that: The specific process of transmitting the data collected by detection to the data fusion processing system for processing. When it is detected that there are hidden dangers in the operation area of the target dike section, output the coordinate of the hidden danger position through the multi-modal navigation system, and generate a hidden danger distribution map is as follows: Collect the ground penetrating radar data and transient electromagnetic data of the geological structure detection, and preprocess the collected ground penetrating radar data and transient electromagnetic data of the geological structure detection; Set the initial resistivity according to prior information or experience and dielectric constant distribution, and construct a forward model based on the initial resistivity and dielectric constant distribution; According to the resistivity in the forward model Generate forward transient electromagnetic response data; According to the dielectric constant in the forward model generate forward ground penetrating radar response data; Compare the forward transient electromagnetic response data with the transient electromagnetic data collected for geological structure detection, and calculate the fitting difference of the electromagnetic data ; Compare the forward GPR response data with the GPR data collected for geological structure detection, and calculate the fitting error of the GPR data ; According to the current resistivity and dielectric constant distribution, calculate the cross-gradient constraint term : ； In the formula, is the gradient operator; Combine , and to construct an objective function for a joint inversion model based on the Bayesian framework : ； In the formula, represents model parameters, including resistivity and dielectric constant ; α, β, and λ are the weight coefficients of each item; Using the gradient descent method, by minimizing the objective function to optimize the inversion model parameters and iteratively update the inversion model parameters: ； In the formula, represents the model parameter vector after the (k + 1)-th iteration; represents the learning rate; After each iteration, check whether the change rate of the objective function is less than the preset change rate or whether the number of iterations has reached the preset maximum value. If so, stop the iteration; otherwise, continue to iterate and update the model parameters; Combine the optimized model parameters with the digital elevation model to generate a hidden danger distribution map and perform three-dimensional visualization annotation on the hidden danger areas.

Images